Image feature point extraction method and device, computer terminal and storage medium

Abstract

The embodiment of the invention discloses an image feature point extraction method and device, a computer terminal and a storage medium. The method comprises the following steps: acquiring a two-dimensional visible light image as a first training sample image; performing homography matrix transformation on the first training sample image to obtain a second training sample image; inputting the first training sample image and the second training sample image into a preset convolutional network model, putting obtained output into a constructed loss function to train parameters of the convolutional network model, and finally obtaining an image feature point extraction model; and inputting a to-be-recognized image into the trained image feature point extraction model to obtain a target feature point.

Description

technical field [0001] The invention relates to the field of machine learning, in particular to an image feature point extraction method, device, computer terminal and storage medium. Background technique [0002] The existing feature point extraction methods generally adopt the following schemes: 1. Use traditional image features, such as harris, sift, surf, etc.; 2. Mark feature points by manual calibration, and then use deep learning methods to train ;3. Use traditional image methods to detect feature points, and use deep learning methods as labeling information for training; 4. Use some 3D or 2D drawing methods to generate some simple pattern images. Because they are generated under control, some basic features can be obtained points, and then trained using deep learning methods. [0003] The existing methods have the following disadvantages: 1. The traditional feature extraction method is effective, but the adaptability is average. It is often necessary to select some ...

Images